What is happening with this loopback reference?

[DaveBode]

Super impressed with REW! I am working on finalizing treatment in my studio and this is the latest measurement with loopback reference. It looks nuts.

When I apply the estimated IR Delay (which is correct), I get this.

That's the proper procedure, correct?

Also, how does this look?

 

[John Mulcahy]

The phase response isn't a particular concern when looking at the room response and effects of treatment, don't need a timing reference for that.

You have a pretty extreme axis range, making things look artificially flat. Best having a 60 dB span - for example, 45 dB at the bottom and 105 dB at the top. You can set the range using the Limits button at the top right of the graph.

The waterfall looks well behaved above 100 Hz, but the frequency axis is linear rather than log which makes it look a little odd - click the Freq. Axis button to toggle between linear and log.

 

[DaveBode]

Thanks, John.

Here is an updated SPL and Waterfall snapshot post-EQ.

 

[DaveBode]

Whoops, I think this is the waterfall in log format

 

[John Mulcahy]

Looks good. The main response looks too flat

If you post the mdat file for the measurement happy to check it over.

 

[DaveBode]

I'd be very happy if you checked out the measurement!

Thanks!

 

[DaveBode]

Not sure what too flat means. I work in video/audio post-producion so I figured this is a good place to start. I might like a bit more in the 30-85 range, but this is coming from a pair of powered studio monitors (Behringer Truth 2031A).

 

[John Mulcahy]

Measurement looks good. Seems like a heavily treated space, pretty dead.

By 'too flat' I was wondering whether a lot of EQ had been applied to flatten it out. EQ above the bass range (a couple of hundred Hz or so in domestic-sized rooms) can easily do more harm than good, unless it is correcting a known deficiency in the speaker's anechoic response. That response looks perhaps 2 or 3 dB high above about 3 kHz.

 

[DaveBode]

Thanks for taking a look at it. There is a lot of treatment in the room. It's a small home studio space where I work on video post-prod and music. Roughly 11x11x8.75'.

I have 12 broadband absorption panels covering the RFZ and back and front walls. All but 2 are 24x48" using 2" thick Roxul Rockboard 80 (compressed mineral wool). And the other 2 are 34"x48" using 3" Roxul Safe n Sound. Those Safe and Sound panels were my first gen units.

In the front of the room (behind my desk) I have 3 edge absorber "bass traps". They are the same 24x48" using 2" mineral wool, but positioned in the corners so they have a larger air gap. Right right and left corners are 2 panels high with a 6" riser, so they go just about to the ceiling. The other front "bass trap" is 17x96" using 3" Safe n Sound. That one is positioned where the front wall and ceiling meet. I have one more corner edge absorber "bass trap" (24x96" 2" Rockboard) where the back wall meets the ceiling.

The "bass traps" were an experiment from something I saw on arqen.com. I have also read or heard Ethan Winer (Real Traps) talk about the same idea. Because the higher traps are so large and such a huge pain in the to mount, I haven't done a test with them in and out of the room. I have tested one of the corner traps in and out of the room and could see some improvement, but it's really small. Based on that test, I am guessing that they are doing a little absorption below 100Hz, perhaps down to 60Hz. But each trap might only be 1-2dB in that range.

I have 2 more 17x48" 3" safe n sound panels just below the monitors tied to the monitor stands. I think it is 22 panels in various arrangements. I made the bass traps because I had a bunch of broadband absorbers in another space that I wasn't using. EQ Below. I am going to be making some more measurements and tweaks in the next few weeks. This EQ was using an average of a couple measurements in the listing position. Something like a few inches in front and behind where I stand/sit (standing desk). I didn't bother to do right and left or the listening position. It get's gnarly real quick.

Equaliser: DCX2496
A,B average
Filter 1: ON BP Fc 2510 Hz Gain -4.5 dB Q 2.00
Filter 2: ON BP Fc 1210 Hz Gain -3.9 dB Q 4.00
Filter 3: ON BP Fc 634 Hz Gain -9.2 dB Q 7.90
Filter 4: ON BP Fc 510 Hz Gain 7.2 dB Q 1.60
Filter 5: ON BP Fc 438 Hz Gain -7.7 dB Q 10.00
Filter 6: ON BP Fc 266 Hz Gain -9.8 dB Q 1.80
Filter 7: ON BP Fc 124 Hz Gain -9.4 dB Q 6.30
Filter 8: ON BP Fc 104 Hz Gain 4.6 dB Q 8.90
Filter 9: ON BP Fc 73.7 Hz Gain 1.2 dB Q 2.20

 

[John Mulcahy]

That is quite a bit of filtering. Have you tried measuring in a few positions around your usual spot? Worth checking what sort of area those settings work for, response can vary a lot over quite short distances.

 

[DaveBode]

I will do some more tweaks this weekend. This was using measurements 6" forward and aft of my listening spot. It's a desk so I don't really go anywhere outside of that.

 

[DaveBode]

Well, that was a long weekend.... I got bogged down with work projects but I was able to tinker do more measurements today. I added an additional panel on the ceiling and did a bunch more measurements. I measured left and right of center and mostly individual speakers for a total of 27 measurements. I generated EQ filters using only the individual left and right measurements and using the separate left and right combined with measurements taken from both speakers at the same time. The filters were very similar.

The filters were a good bit more conservative this time around.

Avg of sep L+R
Filter 1: ON BP Fc 316 Hz Gain -5.5 dB Q 1.80
Filter 2: ON BP Fc 124 Hz Gain -8.6 dB Q 7.90
Filter 3: ON BP Fc 214 Hz Gain 9.1 dB Q 8.90
Filter 4: ON BP Fc 192 Hz Gain -8.1 dB Q 7.90
Filter 5: ON BP Fc 2620 Hz Gain -2.2 dB Q 2.00
Filter 6: ON BP Fc 233 Hz Gain -5.9 dB Q 10.00
Filter 7: ON BP Fc 499 Hz Gain 2.5 dB Q 4.00
Filter 8: ON BP Fc 634 Hz Gain -3.7 dB Q 10.00
Filter 9: ON None

 

[John Mulcahy]

A lot of measurements indeed! The filter at 634 Hz might be a bit narrow being that high in frequency, but generally looks good. Curious how everything looks a bit shelved down below 100 Hz and that there is a hump from around 100 - 500 Hz. I came across this while looking for info on the B2031A by the way, worth a look: http://noaudiophile.com/Behringer_Truth_B2031A/.

 

[DaveBode]

Thanks, John. I had also read that article from noaudiophile. I really like that site and his reviews as it helps to see through some of the marketing when you see the guts of speakers and what components they use in the electronics. I just picked up a second pair of used B2031A monitors for my wife's keyboard this weekend. I was using the Micca PB42x powered speakers, also reviewed on noaudiophile, but they didn't seem to handle the louder synth sounds of the keyboards... Or more likely my kids hammering the snot out of the synth sounds. One of the woofers makes a lovely crackly sound.

I added a shelf filter around 80Hz to bring up that up. It feels a bit better now. I am going to let it simmer for a little while and revisit in a few weeks.

I really appreciate your thoughts and you taking the time to look at the measurements. Thank you very much!